On Simultaneous Interval Estimating the Relative Prevalence of Forward Shifting in Reported Regular Smoking Initiation Age

• Activity Number: 444
• Type: Contributed
• Date/Time: Tuesday, August 6, 2013 : 2:00 PM to 3:50 PM
• Sponsor: Survey Research Methods Section
• Author(s): Brianna Bright*+ and Julia Soulakova
• Companies: University of Nebraska-Lincoln and University of Nebraska-Lincoln
• Keywords: multiple testing ; survey methodology ; tobacco ; simultaneous confidence intervals

Abstract

We discuss an application of Hsu & Berger's (1999) stepwise method for constructing simultaneous confidence intervals for the differences of two binomial proportions, where comparisons to a common control are of interest. While the original method was presented in the context of determining minimum effective dose in drug trials in terms of normally distributed data, the procedure is well suited for any model where the parameters monotonically increase or decrease over some specified groups. Applications of the method to survey data with binary outcomes are discussed with a focus on examining the relative prevalence of forward shifting error. First, logically ordered objectives are defined in the context of comparing forward shifting across initiation age of regular smoking. Next, the 2002-2003 Tobacco Use Supplement to the Current Population Survey data are used to construct the simultaneous confidence intervals adjusted for the multi-stage sampling design and other covariates. Simultaneous confidence intervals will also be constructed using the Bonferroni method for comparison. The key results as well as advantages and disadvantages of the approach are addressed.